Forest Investment Account (FIA) - Forest Science Program
FIA Project M075023

    Rating options for postattack cutting on affected stands
Project lead: Larson, Bruce
Author: Larson, Bruce C.
Imprint: [Vancouver, BC] : University of British Columbia, 2007
Subject: Forest Investment Account (FIA), Dendroctonus Ponderosae, British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
A major problem for forest managers is the decision of what to do after MPB attack, especially in mixed species stands. There are several choices. A simple option is to clearcut the stand. There are several salvaging options ranging from only harvesting the dead trees to expanding the cut, taking advantage of the already expended fixed costs of logging, by thinning or some type of partial cutting of live trees as well. There is another simple option of taking no action and absorbing the volume loss of the mortality without any type of recovery. Before larger, landscape level, scales can be considered, the decision must be addressed at the stand level. The simplest way of doing this is through an optimization process where the optimal solution for the stand is determined. This is a deterministic process. However, there are many assumptions in this type of analysis and many risks associated with these assumptions. One source of risk is that the biological assumptions might not be accurate. A second source is the stochastic nature of events so that low probability events might dramatically change these assumptions. Another aspect of the decision-making is the question of what metric or combination of metrics to use. The most common method is to consider options merely using cost/benefit or net benefit analysis. In this method options are merely ranked in terms of profitability. Another metric is net present value (NPV). NPV can be thought of as simply a way of comparing the relative value of the future compared to the present. Choice of a high discount rate places a much higher value on the present compared to the future, a medium discount rate places a higher relative value on the future, and choosing zero places equal weight on all times, future and present. A third aspect of the decision is long-term fibre production. The solutions could be evaluated based on MAI. This metric removes the economical value component to the decision. The Crown is especially interested in the decision impacts on long term AAC, however, the impact on MAI will also affect the companies directly both in terms setting of annual harvest and capital investment analysis (such as mill construction and upgrade). Information needed to make these decisions: * Present stand volume and value * Future stand volume and value for each decision option * Risk that assumptions are not valid * Risk that conditions will change (i.e. specific assumptions were valid, but general situation has changed We propose to study this process at the Alex Fraser Research Forest (AFRF) near Williams Lake, using stands in the SBS and the IDF zones. We will group stand types into a relatively small number (10-15). For each stand type we will assign one of three MPB attack levels (low, medium and high). This will be a total of 30-45 stand conditions. Then we will use a Markov process to place these stands into the 4-5 action categories listed above (clearcut, no action, and 2 or 3 types of salvage/partial cutting). The AFRF is a good place to develop this system because it has already had a large degree of mortality in many different stand conditions. A current and accurate inventory is maintained as well as a history of price data so it will be easy to incorporate current a volumes and values. In the Markov process a transition matrix is constructed using a series of probabilities. We will do this for the three different metrics (cost/benefit, NPV, and MAI). For the immediate (cost/benefit) analysis the process will be a static Markov process because there are no future probabilities to consider (Buongiorno 2001). For the future metrics (NPV and MAI) a dynamic process will have to be used because the probabilities will change over time (Kaya, and Buongiorno. 1987). For the future volumes we will have to 'grow' the stand under the different conditions. For this step we will investigate which computer models work best for the stand types and structures we are using. TIPSY, TASS, PROGNOSISBC, and SORTIE will all be considered. We will also have to use future price structures. We will consult with economic groups here at the Faculty of Forestry to determine possible future price structures and the probabilities associated with these prices. This decision support tool uses a Markov framework to combine the information listed above to present forest managers with the decision information necessary to make a well founded decision. This matrix will also provide the due diligence necessary in our current management regulations. The decisions will be presented in terms of the three main metrics used in forest decision making and address the probabilities of changing conditions. The main advantage of this system is that the decisions are very transparent (no decision making black box) and simple in construct. The goal is to provide the forester with a tool that is quick and easy to use.


Executive Summary (15Kb)
Presentation (1.0Mb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to